Systemic sclerosis (SSc) is a heterogeneous systemic disorder characterized by alterations of the microvasculature, disturbances of the immune system, and massive deposition of collagen and other matrix substances in connective tissue. Sicca symptoms are thought to be a frequent complaint associated with SSc (1), although little is known about how often they occur and the histopathologic characteristics of the glandular lesions that lead to salivary hypofunction (2).
Sjögren's syndrome (SS) is a chronic autoimmune exocrinopathy that occurs as a distinct entity (primary SS) or in association with other rheumatic diseases, such as rheumatoid arthritis or systemic lupus erythematosus (secondary SS). However, only a few studies have shown evidence of SS in patients with SSc (3–7), and the lack of well-defined criteria for diagnosing SS has led to disparate results: the prevalence of SS in SSc patients has been reported to vary from to 1% (7) to 90% (3).
The main objective of this study was to identify evidence of sicca symptoms and SS according to the American–European Consensus Group criteria (8) in a series of unselected patients with SSc recruited prospectively from 2 rheumatology centers in France. We also sought to identify the disease phenotype association between SSc patients with and those without sicca symptoms and between SSc patients with and those without SS.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
All consecutive SSc patients referred over a 12-month period (from October 1, 2004 to October 1, 2005) to the Department of Rheumatology A, Cochin Hospital in Paris and to the Hautepierre Hospital in Strasbourg were included in the study. We performed a global evaluation of these patients, and collected the following data: age, sex, cutaneous SSc subtype as defined by the LeRoy criteria for SSc subsets (9), disease duration (date of first non–Raynaud's phenomenon symptom), duration of Raynaud's phenomenon, digital ulceration, and use of prostacyclin. The French version of the Health Assessment Questionnaire in SSc was used to evaluate functional disability (10). Systemic assessment was performed as recommended in the consensus core set of variables for SSc studies (11). Computed tomography (CT) scans and respiratory function tests (forced vital capacity and the diffusing capacity for carbon monoxide divided by alveolar volume) were used to assess pulmonary fibrosis. Pulmonary arterial hypertension was defined as pulmonary arterial pressure >40 mm Hg with the patient at rest, as assessed by Doppler echocardiography (12). The following biologic tests were performed: routine blood tests and serologic tests for antinuclear and anticentromere antibodies (by immunofluorescence on HEp-2 cells), anti–topoisomerase I (anti–topo I), anti-RNP, and anti-Ro/SSA or anti-La/SSB antibodies (by counterimmunoelectrophoresis and/or immunodiffusion), and rheumatoid factor (by enzyme-linked immunosorbent assay [ELISA]).
The systematic first-line clinical evaluation included a questionnaire specific for the subjective presence of xerophthalmia and xerostomia, based on the revised American–European Consensus Group criteria for SS (8), together with the Schirmer I test. Findings on the Schirmer I test were considered abnormal if ≤5 mm of the paper was wet after 5 minutes. If the findings of the first-line clinical evaluation were positive (presence of either xerophthalmia or xerostomia or abnormal findings on the Schirmer I test), a labial salivary gland biopsy was performed under local anesthesia.
Tissue specimens included at least 4 salivary glands. Minor salivary glands (obtained through normal-appearing mucosa) were evaluated by an expert pathologist (CD or MCV-L) who was blinded to the diagnosis. The presence of focal lymphocytic sialadenitis with a focus score ≥1 (grade 3 or 4 on the Chisholm and Mason scale, assessing the number of lymphocytic foci/4-mm2 area of salivary gland tissue ) was necessary for a diagnosis of SS (8). Salivary scintigraphic analysis was not available for this study. The presence of sicca syndrome, the SS-specific antibodies anti-Ro/SSA or anti-La/SSB, and/or a focus score ≥1 on labial salivary gland biopsy were considered indicative of associated SS, as detailed in Table 1 (8).
Table 1. Criteria for associated Sjögren's syndrome in patients with systemic sclerosis*
|1. Xerophthalmia (by patient questionnaire)|
|2. Xerostomia (by patient questionnaire)|
|3. Abnormal findings on the Schirmer I test (≤5 mm of wetting in 5 minutes)|
|4. Positive findings on labial salivary gland biopsy (grade 3 or 4 on the Chisholm and Mason scale)|
|5. Presence of anti-Ro/SSA or anti-La/SSB antibodies (by immunodiffusion)|
Glandular fibrosis was defined as the presence of fibrous tissue with secreting fibroblasts surrounding the acini and located around capillaries and excretory ducts. Fibrotic lesions were divided into 3 patterns of severity (mild, moderate, and severe) according to the findings of the histologic analysis.
The study was conducted in accordance with the recommendations of the Declaration of Helsinki. All investigations were those routinely required to evaluate the patients. Each patient gave informed consent for all procedures.
Data were statistically analyzed using chi-square tests with Yates' correction, nonparametric tests, and multivariate logistic regression analysis.
- Top of page
- PATIENTS AND METHODS
Our findings from this prospective and exhaustive series show a marked prevalence of sicca symptoms among SSc patients, who can be divided into 2 groups according to the pathologic process: those with extension of SSc-related fibrosis and those with inflammation in the context of associated Sjögren's syndrome. Of the 133 patients, 19 (14%) had SSc-associated SS as defined by the American–European Consensus Group criteria for SS (8). SS was considerably associated with the limited cutaneous subtype of SSc and with a lower prevalence of pulmonary lesions. Analysis of disease phenotype associations showed that glandular fibrosis was not associated with characteristics of SSc.
From our study results, the principal cause of sicca syndrome in SSc appears to be glandular fibrosis, rather than SS lymphocytic sialadenitis. Indeed, of the 78 biopsy specimens obtained from patients with subjective symptoms of xerostomia, 45 revealed glandular fibrosis and 18 showed lymphocytic sialadenitis, meaning that 23% of the xerostomia was explained by SS and 58% was explained by glandular fibrosis. This observation is not consistent with most previous studies, in which sicca syndrome appeared to be mainly secondary to SS and not to SSc-related fibrosis of the salivary glands (1, 3, 14). However, those studies primarily considered selected SSc patients, whereas we performed a systematic screening that may account for the discrepancies.
Our results are consistent with those of a previous histologic study in which labial salivary gland biopsy samples were obtained from 33 SSc patients, and in which histologic analysis revealed 27 cases of glandular fibrosis and 5 cases of both glandular fibrosis and a focus score ≥1 (2). Another study by the same group of investigators (15) found that whole unstimulated salivary flow was correlated with fibrosis, but not with lymphocytic infiltration, suggesting that xerostomia in SSc was attributable more to the primary sclerosing process of the disease than to the lymphocytic infiltration of an associated SS.
The significant role of fibrosis in xerostomia may be explained by fibrosis occurring around capillaries and excretory ducts (2). Capillary wall sclerosis could induce functional abnormalities by reducing vascular permeability, and periductal fibrosis could interfere with salivary excretion (15). Therefore, sicca syndrome in SSc is probably a complication of the fibrotic process, which is the hallmark of SSc, although this association could not be formally demonstrated in our study because the SSc patients without sicca syndrome did not undergo labial salivary gland biopsy. The consequences of this should be investigated, and stomatologic studies in SSc patients are warranted. It has been suggested that vascular lesions may possibly play a role (16–18), but we did not evaluate this in our study, since we focused only on fibrotic and inflammatory processes. This possibility warrants further study. Pilocarpine has been found to cure oral dryness and keratoconjunctivitis sicca in patients with SS (19, 20), and although the cause of glandular dysfunction is rather distinct in SSc, the high prevalence of sicca symptoms should lead to an evaluation of this drug in SSc patients.
One aim of our study was to clarify the wide range of prevalence of SSc-associated SS that has previously been reported (3–5, 21–23) and which can be explained by the heterogeneity of the criteria used to diagnose SS. Our present study is the first to apply the American–European Consensus Group criteria for SS to a population of patients with SSc. Sjögren's syndrome was diagnosed in 14% of our patients, which is a lower frequency than that reported in previous studies (∼20%) (1, 3–6, 21–23) and is close to the rate of secondary SS reported in rheumatoid arthritis (9–15%) and systemic lupus erythematosus (12%) (24). However, the fact that salivary gland involvement (item V in reference 8) was not available for our study led us to use a variation of the American–European Consensus Group criteria that may have underestimated the frequency of SS. This requires further investigation that takes into account the fact that fibrotic involvement of the salivary glands in SSc may also influence these parameters (15).
The prevalence of anti-Ro/SSA antibodies in our series was lower than that in previous series. Four of the 19 patients with SSc-associated SS (21%) were positive for anti-Ro/SSA antibodies. In contrast, Drosos et al (23) and Osial et al (5) found that anti-Ro/SSA antibodies were present in 33% (3 of 9) and 29% (5 of 17), respectively, of their patients with concurrent SSc and SS. Bell et al (25) found that 60% of SSc patients with sicca syndrome in their series were positive for anti-Ro/SSA antibodies. However, our techniques for detecting anti-Ro/SSA antibodies (immunodiffusion) may partly explain this low frequency, since it is less sensitive than an ELISA method. Nevertheless, our results are consistent with those of Akiyama et al (26), who found that anti-Ro/SSA and anti-La/SSB antibodies were less common in patients with secondary SS associated with lcSSc than in patients with primary SS. A subgroup of patients with primary SS and anticentromere antibodies, but with a low prevalence of anti-Ro/SSA antibodies, has previously been described (27). Our data suggest that SS in SSc resembles secondary SS (28, 29) rather than primary SS, in which the presence of anti-Ro/SSA antibodies is far more common.
Our data show that SS is much more common in patients with the limited cutaneous form of SSc who have anticentromere antibodies, and it is associated with a lower frequency of pulmonary fibrosis. The association between SS and the presence of anticentromere antibodies can be explained by the higher prevalence of SS in patients with lcSSc (9, 14, 30, 31). Miyawaki et al (14) found that 84 of their series of 120 patients positive for anticentromere antibodies had lcSSc. Of these patients, 25 (30%) had overlap with SS, 4 had overlap with primary biliary cirrhosis, and 4 had overlap with rheumatoid arthritis, thus highlighting the association between lcSSc and autoimmune-related diseases or autoimmune phenomena. In this study, we did not assess overlap between SS and other autoimmune diseases, such as primary biliary cirrhosis or autoimmune thyroiditis, and this should be investigated further.
The propensity of SSc to cause autoimmune effects is thought to be due to the production of Th2 cytokines and the activation of B cells in relation to the expression of systemic autoimmune diseases (32, 33). Patients with SSc have altered B cell homeostasis, which is characterized by expanded numbers of naive B cells and diminished numbers of memory B cells, and they have high numbers of CD4+,CD30+ T cells in the skin and increased serum levels of soluble CD30, suggesting the possible involvement of Th2 (34–38). This may be particularly true for patients with lcSSc, in whom there is an increased production of interleukin-10 (38–41) and in whom clonally expanded T cells are more frequent as compared with patients with dcSSc (42). Taken together, these findings suggest that the limited cutaneous subtype of SSc carries a greater autoimmune background, and this may have implications in patients for whom there is no specific treatment (43); however, further study is required.
SS can cause specific systemic complications. Although concurrent SS-related complications were not specifically evaluated in our study, routine screening did not show any tubulointerstitial nephritis or neuropathy, which is consistent with previously reported data (25). Although age and disease duration did not influence the link between sicca syndrome and associated SS or labial salivary gland fibrosis, more data are needed to evaluate the variations and relationship between labial salivary gland inflammation and fibrosis over time.
In summary, in this series of consecutive patients with SSc, the prevalence of sicca syndrome and SSc-associated SS was 68% and 14%, respectively. The main cause of sicca syndrome appears to be fibrosis, and involvement of the exocrine glands may be considered a complication of the fibrotic process that is the hallmark of SSc. SS predominates in patients with lcSSc, which is thus suspected to be associated with a specific B cell–mediated autoimmunity and some phenomena of autoimmune dysfunction that are different from those associated with dcSSc.